Luis Pater ecc9aa72b3 fix(openai): preserve assistant content when converting Responses tool-call turns ...

- Merge buffered `tool_calls` into the latest mergeable assistant message instead of always appending a new one.
- Combine deferred reasoning segments across the same assistant turn, while ignoring `[reasoning unavailable]` placeholders.
- Reset merge state on role/tool-output boundaries to keep tool-call attachment behavior correct across turns.

Closes: #4676

2026-08-10 05:13:24 +08:00

..

antigravity

Reduce Antigravity request amplification for large payloads

2026-08-09 00:40:57 +08:00

claude

fix(openai,claude): normalize Responses custom tool-call identity with shared namespace handling

2026-08-09 15:30:13 +08:00

codex

Reduce Codex request amplification for large payloads

2026-08-07 19:16:06 +08:00

common

fix(openai,claude): normalize Responses custom tool-call identity with shared namespace handling

2026-08-09 15:30:13 +08:00

gemini

Make the payload-reuse guards able to fail

2026-08-09 00:40:57 +08:00

interactions

feat(compat): preserve compat-mode thinking/signature blocks for API-key models

2026-08-06 17:19:24 +08:00

openai

fix(openai): preserve assistant content when converting Responses tool-call turns

2026-08-10 05:13:24 +08:00

translator

chore: upgrade CLIProxyAPI dependency to v7 across the project

2026-05-08 11:46:46 +08:00

init.go

feat(interactions): add support for Google Interactions

2026-07-07 02:39:13 +08:00

request_benchmark_test.go

perf(translator): add unit tests for payload reuse and normalization across translators

2026-07-20 03:50:35 +08:00

response_benchmark_test.go

perf(translator): add unit tests for payload reuse and normalization across translators

2026-07-20 03:50:35 +08:00